Filters - Chokes - Opto-isolators
in model radio installation.
This is an extract from a thread on the forum started due to the loss of a model at a competition, through an unknown cause. All checks made since have failed to identify the problem. Many competition flights had been made during the weekend including the model in question. The weather was less than ideal with heavy and prolonged rain showers, although it was not raining at the time the problem occurred. We were flying on an active airfield that had been used many times before.
There are many who would disagree with complicating a model with unnecessary 'gadgets' but rely on meticulous attention being paid to the quality, care and installation of equipment, and they have had trouble free flying over many years. However we are always looking for ways of improving the rejection to radio interference and the following emerged from the discussion.
Mel Garlick (GBR/CAA Member) was asked to give us some thoughts on filters, chokes and opto-isolators used in model radio installations.
Mel begins "OK, here are some basic facts, but please donít pick me on technicalities as
Iím trying to make this understandable to the complete layman"!
"Opto-isolators, or couplers, are made up of a light emitting device and a light sensitive device with no electrical connection between the two, just a beam of light. The light emitter is usually an LED. The light sensitive device may be a photodiode, phototransistor, thyristor or a triac etc. Usually found as a sealed component
They are used to isolate low-current control or signal circuitry from transients generated or transmitted by power supply and high-current control circuits.
This means that for our purposes they would prevent electrical ďnoiseĒ and voltage spikes from getting down the wires from the servoís.
They would have no effect on RF signals on or near your frequency.
Examples of noise are the crackles and pops that you get on a radio or the lines that appear on a TV when someone is using a hairdryer or electric drill.
Voltage spikes are short bursts of excess voltage.
I would think that our servoís should be well enough designed and constructed so that an opto-isolator would not normally be needed for IC, but there are benefits if using electric power.
If the unwanted signal or interference is coming down the aerial wire, as RF, then there would be no benefit at all, but Iíll go into that subject later".
Types of Ferrite and how they are used to make chokes. (Photos Mel Garlick)
Bead Filter Ring filter Rod filter
"Their primary purpose is to attenuate (reduce) electromagnetic interference and radio frequency interference. Usually known as EMI and RFI
A ferrite choke is a cylinder or ring made of ferrite. This is a semi-magnetic material consisting of iron oxide (yes I know thatís rust) and alloyed with other metals.
With long servo leads, wires or any other conductive material (including carbon fibre) particularly if the length makes them resonant close to the wanted frequency (see my other post) they will act as perfectly good aerials for the signals they carry. They can and will broadcast these signals quite efficiently. The cables can also receive signals and transmit them into the receiver directly, where they will also cause problems.
A ferrite ring or bead has the property of eliminating , or at least reducing , these unwanted signals. Essentially, it "chokes" the RFI transmission at the point on the cable where it is inserted. This is why you should fit the ferrites at the receiver end of the cable. Instead of travelling down the cable and transmitting, the RFI signals turn into heat in the ferrite.
High frequency, thatís about 50 MHz and higher, currents are likely reduced more by single turn ferrite chokes or beads. whereas low frequency signals, below 10 MHz, are definitely reduced more by multiple turn ferrite chokes. The exact frequencies depend on the ferrite used and impedances where the ferrites are used.
Our frequencies come somewhere between the two so no particular advantage appears from using either type over the other. There is no single grade of ferrite which is ideal for use on all frequencies. The Fair-Rite type 43 material gives good results from 7MHz upwards though.
Normally, any RF signal of 10dB or more less than the wanted signal on the same or close frequency will be ignored by an FM receiver. Any signal stronger than this is potentially NOT going to be ignored, resulting in glitches or even a total lockout if the signal is strong enough. 10dB is roughly 1/10th the strength of 0dB. Every 3dB either halves or doubles, depends which way you are going.
It doesnít have to be as strong as your wanted signal to cause severe problems"!
"If you are trying to prevent unwanted RF signals as opposed to noise there are various types of filter that you can use. The receiver itself will include some of these
They are likely to involve circuitry which could include capacitors inductors and resistors.
Band pass, low pass, high pass and notch are all we need to know for simplicity.
Low pass filters will pass low frequency signals but severely reduce high frequency signals.
High pass will pass high frequencies but reduce lower ones..
Notch will block a specific frequency
Band pass will allow a narrow band of frequencies but block higher and lower.
High and low pass are not really applicable to our use. Notch is only useful if you are getting a problem from a frequency close to your own. If I was on a site that had that problem I wouldnít fly there anyway!
This leaves us with bandpass. A band-pass filter works to screen out frequencies that are too low or too high, giving easy passage only to frequencies within a certain range. They are a combination of high pass and low pass combined into a single filter. In theory a filter could be designed to allow 35MHz but block others.
In practice it certainly wouldnít be that good but it would reduce unwanted signals from outside the band.
The manufacturers can probably come up with reasons for using their filters, after all they are in the business of selling them. Sometimes there may be a valid reason. You will have to make your own decision
My opinion would be that as long as you donít have a specific problem and use good radio housekeeping. In other words make sure you donít have metal to metal joints, keep conductors away from your aerial, break any possible resonant leads with a simple ferrite ring or bead. (Thatís probably servo leads over about 18 inches to be safe), then you donít need any other filters.
For the more cost conscious out there, a very good way of getting an efficient ferrite choke is a few turns of the servo lead wire (extension lead if necessary?) around a section of ferrite rod aerial salvaged from a transistor radio. Heat shrink it on and you have a ďfreeĒ filter!
If you want to buy one, then the Futaba filter, part number FUTM4149, or similar from the other manufacturers would be suitable.
Just to close, after 25 years of playing about with filters and despite having
some of the most sophisticated filters now available, including digital signal
processing ones, I still canít always filter out unwanted signals to be able to
hear the one that I do want.
Beware of exaggerated claims made for any filter"!
Thank you Mel.
If anyone would like to pursue this subject I am sure a post on the forum
would stimulate more discussion.